Hybrid Kinematic and Muscular Null Space for Simultaneous Control of Natural and Extra Degrees of Freedom.

Humans possess more degrees of freedom than are required to perform upper-limb tasks, giving rise to a body null space that can be exploited to extend natural control capabilities. Both kinematic and muscular null space control modalities have been shown to be feasible in prior studies; however, their relative performance has not yet been compared within the same participants and task. Moreover, their potential combined use for flexible control of supernumerary degrees of freedom (SuDoFs) remains unexplored. In this study, we compare kinematic and muscular null space control and investigate a hybrid modality that either combines both to control one SuDoF or uses them separately to control two SuDoFs simultaneously. We propose a novel implementation of kinematic and muscular null space extraction that enables their real-time integration within a unified control framework. Sixteen participants performed a reaching and orientation task in a virtual reality environment, controlling the rotation of a virtual elliptical cursor to match targets with varying orientations. Task difficulty was modulated through single-axis and dual-axis target rotations, as well as with a baseline condition requiring no rotation control. Participants completed the task using muscular, kinematic, and hybrid null space control modalities. Results showed that muscular control outperformed purely kinematic control for rotated targets. The hybrid modality significantly improved performance relative to kinematic control, mitigating performance degradation, and supported both combined single-SuDoF control and simultaneous two-SuDoF control. These findings demonstrate that hybrid null space control is a robust and flexible strategy for human augmentation.